CN202650529U - Simple and easy voice gender detection apparatus - Google Patents
Simple and easy voice gender detection apparatus Download PDFInfo
- Publication number
- CN202650529U CN202650529U CN 201220299814 CN201220299814U CN202650529U CN 202650529 U CN202650529 U CN 202650529U CN 201220299814 CN201220299814 CN 201220299814 CN 201220299814 U CN201220299814 U CN 201220299814U CN 202650529 U CN202650529 U CN 202650529U
- Authority
- CN
- China
- Prior art keywords
- module
- voice
- signal
- input end
- easy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The utility model discloses a simple and easy voice gender detection apparatus. The simple and easy voice gender detection apparatus comprises a voice acquisition module, a low-pass filtering module, a frequency voltage conversion module, an analogue-to-digital conversion module, a controller module and a display module, wherein the voice acquisition module is used for converting an acoustic signal into electric waves; the low-pass filtering module, whose input end is connected with the voice acquisition module, is used for filtering out higher harmonic waves from the electric waves; the frequency voltage conversion module, whose input end is connected with the output end of the low-pass filtering module, is used for converting a frequency signal into a voltage signal; the analogue-to-digital conversion module, whose input end is connected with the output end of the frequency voltage conversion module, is used for converting an analogue signal into a digital signal; the controller module, whose input end is connected with the output end of the analogue-to-digital conversion module, is used for determining the magnitude of the digital signal; and the input end of the display module is connected with the output end of the controller module. According to the voice gender detection apparatus, the gender distinction of a human voice becomes relatively simple, convenient and fast, training in a mega data amount is prevented, and no computer devices are needed. The simple and easy voice gender detection apparatus not only can serve as a supplement to complex identity authentication devices, but also can be made into a simple and easy hand-help device which can be used conveniently. The simple and easy voice gender detection apparatus has a certain market prospect.
Description
Technical field
The utility model relates to a kind of simple and easy voice sex-screening device.
Background technology
Voice signal is communication modes the most common and commonly used in people's daily life, and the development of modern signal processing technology is so that the speech-sound intelligent recognition technology is ripe all the more.After IBM Corporation released commercial speech recognition software Via Voice, the identification of the computing machine of voice progressively entered daily life, and the voice booking at station, the automatic speech dialing of mobile phone etc. are used and be familiar with by everybody.The technology of utilizing simultaneously voice to carry out authentication also progressively receives publicity, but accuracy is not high, still has a lot of restrictions in the application of some important events.Identification speaker's sex is comparatively basic in an authentication task, and classic method need to gather a large amount of men and women's sound data as Sample Storehouse, and sets up complicated HMM model of cognition, trains with Sample Storehouse, thereby sets up the recognition template of men and women's sound.Do like this so that data volume is larger, nor one satisfy the ubiquity of voice surely, recognition performance and the Sample Storehouse degree of correlation are larger; Accuracy and the training utterance of identification in a minute in have larger contacting, well more a lot of than the recognition effect of nonspecific text based on the recognition effect of particular text; Voice signal will produce along with the difference of sound pick-up outfit difference simultaneously, even same person is recorded one section voice at an equipment, also can there be difference in the data recording of same language on another equipment, if recording microphone material Different Effects is then larger.Distinguish equipment based on the voice sex of category of model and all have problems.
Summary of the invention
In order to address the above problem, the purpose of this utility model is to provide a kind of simple and easy voice sex-screening device, by the frequency after the voice signal shaping, mainly relies on the difference of men and women's fundamental frequency, judges speaker's sex.
In order to reach above purpose, the technical solution adopted in the utility model is: simple and easy voice sex-screening device, and it comprises the voice acquisition module that voice signal is transformed into electrical waveform, input end is connected with voice acquisition module and is used for the low-pass filtering module of the higher hamonic wave of filtering electrical waveform, input end is connected with the output terminal of low-pass filtering module and is used for frequency signal is converted to the voltage to frequency modular converter of voltage signal, input end is connected with the output terminal of voltage to frequency modular converter and is used for analog signal conversion is the analog-to-digital conversion module of digital signal, input end is connected with the output terminal of D/A converter module and be used for judges the controller module of digital signal size, the display module that input end is connected with the output terminal of controller module.
Further, low-pass filtering module is second-order filter circuit, and it comprises amplifier, resistance and electric capacity.
Further, the passband of low-pass filtering module is 10～500Hz, is limited to 600Hz under the stopband, and transitional zone decays more than 20dB, and passband fluctuation is in 3dB.
According to different classifying rules Chinese several articulation types can be arranged, wherein whether the vibration with vocal cords is divided into voiceless sound and voiced sound with voice, people's vocal cords will produce vibration when sending out voiced sound, sound wave presents quasi-periodic characteristic, vocal cords do not vibrate substantially when sending out voiceless sound, sound wave presents the characteristic of similar white noise, and rule is not obvious.The frequency of vocal cord vibration is called fundamental frequency, and people's fundamental frequency generally is between 50～400Hz.By a large amount of priori understanding as can be known, the fundamental frequency of general adult male is at 50～200Hz, and adult female's fundamental frequency is distributed in 200～400Hz, and children's fundamental frequency is similar to women's fundamental frequency.Generally speaking, after the shaping of voice signal process low pass, the energy of speech pitch section can preserve preferably.Utilize the detection method of above-mentioned simple and easy voice sex-screening device, it may further comprise the steps: (1) sets the cut-off energy of voiced sound according to neighbourhood noise; Set the boundary frequency of men and women's sound; (2) collected sound signal judges that the energy value of this voice signal whether greater than cut-off energy, if greater than cut-off energy, then enters next step, as less than or equal to cut-off energy, and Resurvey voice signal and judging then; (3) if the energy value of voice signal greater than cut-off energy, is then opened the signalling channel of voltage to frequency modular converter, controller module records one group of analog-to-digital magnitude of voltage, and averaged; (4) whether detect this mean value greater than boundary frequency, be then to be female voice, otherwise be male voice, and control LCD makes result's demonstration; Return subsequently original state, carry out next group data acquisition and process.
The utility model has been avoided the training of big data quantity, and has not been needed computer installation so that distinguish based on the sex of people's voice and to become simple and quick.This device both can be used as replenishing of complicated authentication means, also can make an easy handheld device, and was easy to use, had certain market outlook.
Description of drawings
Fig. 1 is the utility model device circuit structural representation.
Fig. 2 is the detection method process flow diagram of the utility model device.
Embodiment
Below in conjunction with accompanying drawing preferred embodiment of the present utility model is described in detail, thereby so that advantage of the present utility model and feature can be easier to be it will be appreciated by those skilled in the art that protection domain of the present utility model is made more explicit defining.
Can find out from the structural representation of accompanying drawing 1, the utility model provides simple and easy voice sex-screening device, and it comprises the voice acquisition module that voice signal is transformed into electrical waveform, input end is connected with voice acquisition module and is used for the low-pass filtering module of the higher hamonic wave of filtering electrical waveform, input end is connected with the output terminal of low-pass filtering module and is used for frequency signal is converted to the voltage to frequency modular converter of voltage signal, input end is connected with the output terminal of voltage to frequency modular converter and is used for analog signal conversion is the analog-to-digital conversion module of digital signal, input end is connected with the output terminal of D/A converter module and be used for judges the controller module of digital signal size, the display module that input end is connected with the output terminal of controller module.
In the present embodiment, this device is comprised of microphone voice collecting circuit, 500Hz low-pass filter circuit, F/V change-over circuit, A/D change-over circuit, MCU core processor and LCD display based on LM331.This device can use under any non-strong noise environment of indoor and outdoor.Microphone voice collecting circuit cooperates corresponding resistance and electric capacity to consist of by the cylinder microphone, voice signal can be transformed into electrical waveform; The 500Hz low-pass filter consists of second-order filter circuit by Amplifier OP07 commonly used and resistance, electric capacity, and transitional zone is less than 100Hz, and is main with the higher hamonic wave filtering in the voice more than the decay 20dB, also can play certain shaping operation; The F/V change-over circuit mainly is made of the LM331 chip, can realize the conversion of frequency and voltage; The AD converter that the AD conversion can adopt MCU to carry also can adopt external high-speed AD chip, and switching time, the 10bit quantization digit satisfied matching requirements fully less than 2us; MCU adopts the PIC16F877 single-chip microcomputer, and this device programming conveniently has more than 30 general purpose I/O, and enough the input and output of device distribute, and control LCD, key-press input, and the AD input is all very convenient, integrated UART bus on the while sheet, expansion is convenient in convenient and compunication; Adopt conventional 1602LCD display as the output of device, show the result.
Speaker's voice signal becomes voltage signal by Acquisition Circuit by acoustic pressure, again by low-pass filter circuit, with the higher hamonic wave filtering, the accuracy requirement of wave filter is not high especially, passband 10～500Hz, stopband lower limit 600Hz, more than the transitional zone decay 20dB, in the passband fluctuation 3dB.Filtered signal has been equivalent to finish a part of shaping work, is convenient to follow-up F/V conversion.Understand the mode that frequency content in the signal can adopt digital sample and spectrum analysis, but cost is higher, data processing complex.And transform by F/V, frequency signal has originally become voltage signal, only needs general AD to transform the frequency content that just can directly obtain by change in voltage in the original signal, more convenient and quicker.Size by magnitude of voltage is judged the fundamental frequency of voice signal, with predefined boundary frequency relatively, be judged to be female voice greater than the sound of boundary frequency, otherwise be male voice.Show the result by LCD at last.
Fig. 2 is the detection method process flow diagram of the utility model device.Because the pause of people's pronunciation, voice signal has the space; And fundamental frequency only is included in the voiced sound, do not have in the voiceless sound.So must be to without acoustical signal, voiceless sound signal and voiced sound signal clear and definite differentiation being arranged when voice signal is analyzed.Generally speaking, energy comparison noiseless or the voiceless sound signal is little, and the energy of voiced sound much larger than this two, therefore utilize the energy size of voice can navigate to easily turbid voice.In addition because voice signal has the nonlinear feature of non-stationary, so must could obtain preferably result by the method for short-time analysis.Conventional speech analysis means is divided into one group of 20～40ms with voice and processes (the AD picking rate is divided the frame duration much larger than voice on the sheet of PIC16F877, meets the demands fully).General voiced sound can both be kept more than the 200ms, therefore the error of processing and judging for fear of data, this device will be done the speech pitch of obtaining in the 200ms and carry out sex after the average treatment again and judge the mistake that is caused to resist all kinds of factor.
A kind of detection method of utilizing above-mentioned simple and easy voice sex-screening device, it may further comprise the steps:
(1) according to neighbourhood noise, set the cut-off energy E of turbid voice, namely the signal greater than this thresholding is voiced sound; According to prior imformation, set the boundary frequency F of men and women's sound, greater than this frequency be female voice, otherwise be male voice.
(2) after the device operation, collected sound signal, at first detect voice short-time energy, judge that whether the energy value of this voice signal is greater than cut-off energy, if greater than cut-off energy E, then enter next step, as less than or equal to cut-off energy, then Resurvey voice signal and judgement are known to detect the voiced sound signal;
(3) if the energy value of voice signal greater than cut-off energy E, is then opened the F/V signalling channel, MCU records the magnitude of voltage of one group of AD conversion, the filtering jump signal, and power taking flattens average, and obtains the fundamental frequency of corresponding voice by conversion.
(4) whether detect this mean value greater than described boundary frequency F, be then to be female voice, otherwise be male voice, and control LCD makes result's demonstration; Return subsequently original state, carry out next group data acquisition and process.
Above-described embodiment only is explanation technical conceive of the present utility model and characteristics, and its purpose is to allow the personage who is familiar with technique can understand content of the present utility model and according to this enforcement, can not limit protection domain of the present utility model with this.All equivalences of doing according to the utility model Spirit Essence change or modify, and all should be encompassed within the protection domain of the present utility model.
Claims (3)
1. simple and easy voice sex-screening device, it is characterized in that: it comprises the voice acquisition module that voice signal is transformed into electrical waveform, input end is connected with described voice acquisition module and is used for the low-pass filtering module of the higher hamonic wave of the described electrical waveform of filtering, input end is connected with the output terminal of described low-pass filtering module and is used for frequency signal is converted to the voltage to frequency modular converter of voltage signal, input end is connected with the output terminal of described voltage to frequency modular converter and is used for analog signal conversion is the analog-to-digital conversion module of digital signal, input end is connected with the output terminal of described D/A converter module and is used for judging the controller module of described digital signal size, the display module that input end is connected with the output terminal of described controller module.
2. simple and easy voice sex-screening device according to claim 1, it is characterized in that: described low-pass filtering module is second-order filter circuit, and it comprises amplifier, resistance and electric capacity.
3. simple and easy voice sex-screening device according to claim 1 and 2, it is characterized in that: the passband of described low-pass filtering module is 10～500Hz, is limited to 600Hz under the stopband, and transitional zone decays more than 20dB, and passband fluctuation is in 3dB.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201220299814 CN202650529U (en) | 2012-06-26 | 2012-06-26 | Simple and easy voice gender detection apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201220299814 CN202650529U (en) | 2012-06-26 | 2012-06-26 | Simple and easy voice gender detection apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN202650529U true CN202650529U (en) | 2013-01-02 |
Family
ID=47419597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201220299814 Expired - Fee Related CN202650529U (en) | 2012-06-26 | 2012-06-26 | Simple and easy voice gender detection apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN202650529U (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102693721A (en) * | 2012-06-26 | 2012-09-26 | 苏州科技学院 | Simple and easy voice and gender detection device and method |
| CN103167174A (en) * | 2013-02-25 | 2013-06-19 | 广东欧珀移动通信有限公司 | Output method, device and mobile terminal of mobile terminal greetings |
| CN105679311A (en) * | 2016-02-03 | 2016-06-15 | 郑州职业技术学院 | Automatic opening single-leaf door |
-
2012
- 2012-06-26 CN CN 201220299814 patent/CN202650529U/en not_active Expired - Fee Related
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102693721A (en) * | 2012-06-26 | 2012-09-26 | 苏州科技学院 | Simple and easy voice and gender detection device and method |
| CN103167174A (en) * | 2013-02-25 | 2013-06-19 | 广东欧珀移动通信有限公司 | Output method, device and mobile terminal of mobile terminal greetings |
| CN105679311A (en) * | 2016-02-03 | 2016-06-15 | 郑州职业技术学院 | Automatic opening single-leaf door |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6691090B1 (en) | Speech recognition system including dimensionality reduction of baseband frequency signals | |
| CN103280220B (en) | A kind of real-time recognition method for baby cry | |
| CN104867495B (en) | Sound recognition apparatus and method of operating the same | |
| CN108172242B (en) | Improved Bluetooth intelligent cloud sound box voice interaction endpoint detection method | |
| CN103578468B (en) | The method of adjustment and electronic equipment of a kind of confidence coefficient threshold of voice recognition | |
| CN105206271A (en) | Intelligent equipment voice wake-up method and system for realizing method | |
| CN101424924A (en) | Sound control intelligent household control system | |
| CN102723078A (en) | Emotion speech recognition method based on natural language comprehension | |
| CN103646649A (en) | High-efficiency voice detecting method | |
| CN110992932B (en) | Self-learning voice control method, system and storage medium | |
| CN102693721A (en) | Simple and easy voice and gender detection device and method | |
| CN109949808A (en) | The speech recognition appliance control system and method for compatible mandarin and dialect | |
| CN111667838B (en) | Low-power-consumption analog domain feature vector extraction method for voiceprint recognition | |
| CN105118522A (en) | Noise detection method and device | |
| CN106098078A (en) | A kind of audio recognition method that may filter that speaker noise and system thereof | |
| CN103198829A (en) | Method, device and equipment of reducing interior noise and improving voice recognition rate | |
| CN110136709A (en) | Audio recognition method and video conferencing system based on speech recognition | |
| Reddy et al. | A comparison of cepstral features in the detection of pathological voices by varying the input and filterbank of the cepstrum computation | |
| CN201181413Y (en) | Acoustic control intelligent household control device | |
| CN103139395A (en) | Mobile terminal and control method thereof | |
| CN106033669B (en) | Audio recognition method and device | |
| CN202650529U (en) | Simple and easy voice gender detection apparatus | |
| CN113539294A (en) | Method for collecting and identifying sound of abnormal state of live pig | |
| CN108364641A (en) | A kind of speech emotional characteristic extraction method based on the estimation of long time frame ambient noise | |
| CN101930733A (en) | Speech emotional characteristic extraction method for speech emotion recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130102 Termination date: 20130626 |